Fig. 5
From: MERIT reveals the impact of genomic context on sequencing error rate in ultra-deep applications
Context-specific optimal sequencing depth. Substitution error rates are classified based on their type (column a) and context (columns b and c) at nine different depths: 1,000 ×, 5,000 ×, 10,000 ×, 25,0000 ×, 50,000 ×, 100,000 ×, 200,000 ×, 400,000 ×, and 700,000 ×. In silico depth reduction procedure was performed on merged reads, amplified by polymerase Ultra II to an average depth of 1,930,473 ×. The shaded areas are uncertainty bounds of one standard deviation around the average, derived from 500 independent sub-samples